Participation of 14-3-3ε and 14-3-3ζ proteins in the phagocytosis, component of cellular immune response, in Aedes mosquito cell lines

Background Better knowledge of the innate immune system of insects will improve our understanding of mosquitoes as potential vectors of diverse pathogens. The ubiquitously expressed 14-3-3 protein family is evolutionarily conserved from yeast to mammals, and at least two isoforms of 14-3-3, the ε and ζ, have been identified in insects. These proteins have been shown to participate in both humoral and cellular immune responses in Drosophila. As mosquitoes of the genus Aedes are the primary vectors for arboviruses, causing several diseases such as dengue fever, yellow fever, Zika and chikungunya fevers, cell lines derived from these mosquitoes, Aag-2 from Aedes aegypti and C6/36 HT from Aedes albopictus, are currently used to study the insect immune system. Here, we investigated the role of 14-3-3 proteins (ε and ζ isoform) in phagocytosis, the main cellular immune responses executed by the insects, using Aedes spp. cell lines. Results We evaluated the mRNA and protein expression of 14-3-3ε and 14-3-3ζ in C6/36 HT and Aag-2 cells, and demonstrated that both proteins were localised in the cytoplasm. Further, in C6/36 HT cells treated with a 14-3-3 specific inhibitor we observed a notable modification of cell morphology with filopodia-like structure caused through cytoskeleton reorganisation (co-localization of 14-3-3 proteins with F-actin), more importantly the decrease in Salmonella typhimurium, Staphylococcus aureus and E. coli phagocytosis and reduction in phagolysosome formation. Additionally, silencing of 14-3-3ε and 14-3-3ζ expression by mean of specific DsiRNA confirmed the decreased phagocytosis and phagolysosome formation of pHrodo labelled E. coli and S. aureus bacteria by Aag-2 cells. Conclusion The 14-3-3ε and 14-3-3ζ proteins modulate cytoskeletal remodelling, and are essential for phagocytosis of Gram-positive and Gram-negative bacteria in Aedes spp. cell lines. Electronic supplementary material The online version of this article (doi:10.1186/s13071-017-2267-5) contains supplementary material, which is available to authorized users.


Background
Phagocytosis by haemocytes is the main innate immune process to eliminate invading bacteria in insects [1][2][3][4]. This process is initiated with the recognition of the foreign particle, followed by remodelling of the phagocytic cell cytoskeleton and plasma membrane invagination resulting in internalisation of particles into a phagosome [5]. Phagosomes then undergo serial fusions with endosomes and lysosomes (maturation process), to form a fully functional phagolysosome, where degradation and destruction of ingested particles occur [6]. Several intracellular signalling pathways that drive the insect immune system are similar to those of the mammalian innate immune system [7,8]. Phagocytosis, a major innate immune process in insects, is crucial for tissue homoeostasis and remodelling since it is required for clearance of apoptotic bodies during embryogenesis and metamorphosis where tissues are remodelled, and cells under programmed death have to be removed [8].
Aedes spp. mosquitoes are vectors of disease-causing arboviruses such as dengue, yellow fever, chikungunya and Zika [17,18]. In a previous study, we identified two, 14-3-3ε and 14-3-3ζ isoforms in Ae. aegypti, that were highly homologous to the corresponding Drosophila orthologues, suggesting that they may have conserved functional roles in phagocytosis [13]. In this work, we investigated the role of the 14-3-3 isoforms in phagocytosis of Gram-positive and Gram-negative bacteria in the two cell-lines Aag-2 derived from Ae. aegypti and C6/36 HT from Ae. albopictus.

Cell viability study
To evaluate cell viability in the presence of 14-3-3 inhibitor (Antagonist I, 2-5) [21,22], 8 × 10 4 cells were grown in 96 well plates (2500 cells/mm 2 ) until they reached the exponential phase [23]. The first 15 h the cells remain in the Lag phase of growth. Subsequently, they enter the Log phase of growth; we performed the cell viability at 24 h. Subsequently, cells were gently washed with serum-free medium. The cells were incubated 120 min at 34°C with several concentrations of 14-3-3 inhibitor (dissolved in DMSO) (12.5, 25, 50 and 100 μM), DMSO (vehicle used to dissolve the inhibitor) (Sigma-Aldrich, St. Louis, MO, USA) and medium (without inhibitor) in 100 μl of fresh serum free medium. Cells were then incubated with CellTiter96® AQueous One Solution Reagent for 60, 120 and 180 min at 34°C [23,24] (Promega Corporation, Madison, WI, USA) according to the manufacturer's protocol. The experiments were performed in triplicate.

Phagocytosis study
To evaluate phagocytosis, FITC labelled Escherichia coli [25], and Salmonella typhimurium-GFP [26] were incubated with C6/36 HT cells in a 10:1 ratio for 45 min at 34°C or 4°C (negative control). Mosquito cells were detached by scraping and resuspended in ice-cold PBS. Engulfed bacteria recorded by flow cytometry in a Cyan ADP Analyzer (Beckman Coulter, Inc., Indianapolis, IN, USA). The fluorescence intensity was acquired in the FL1 channel (FITC/GFP). To evaluate phagosome formation, we utilised pHrodo green Bio Particles -E. coli or Staphylococcus aureus labelled with the acid sensor pHrodo-succinimidyl ester (pHrodo™ SE Green, Life Technologies). This dye is activated in acid cell compartments such as phagolysosome [27,28]. Lyophilized pHrodo green BioParticles were reconstituted in Uptake Buffer (20 mM HEPES in HBSS Hank's balanced salt solution, pH 7.4, ThermoFisher) at a concentration of 1 mg/ml prior to use, 50 μl of resuspended pHrodo green BioParticles were added to either C6/36 HT cells cultured with 14-3-3 inhibitor or 14-3-3 silenced Aag-2 cells in 24 wells plates for 120 min. Then, cells were detached, single-cell resuspended and subjected to flow cytometric analysis of fluorescence intensity of pHrodo green. The fluorescence intensity was acquired in the FL1/ FL2 channels (pHrodo green-509/540 nm). The cytometry result was expressed as mean fluorescence intensity (MFI), and phagocytosis percentage and phagocytosis percentage of inhibition [25].

Statistical analysis
GraphPad Prism statistical software package (Prism 5.0; Graph-Pad Software, Inc., San Diego, CA) was used to perform statistical analysis. All data were presented as the mean ± SEM (standard error of the mean). The statistical comparisons of data between the different conditions were performed using one-way analysis of variance (ANOVA) or Student's t-test. P-values below 0.05 were considered statistical significance.

Results
Aedes spp. cells express 14-3-3ε and 14-3-3ζ proteins We first evaluated the gene expression of Aeae14-3-3ε and Aeae14-3-3ζ in cell lines derived from Ae. aegypti and Ae. albopictus. RT-PCR amplifying specific segments of 14-3-3ε and 14-3-3ζ transcripts confirmed that they were highly expressed in C6/36 HT and Aag-2 cells (Fig.  1a, b). In addition, Western blot assays using anti-14-3-3 antibody (clone K-19) demonstrated two bands with the predicted molecular weights of~30 and~29 kDa corresponding to 14-3-3ε and 14-3-3ζ protein isoforms, respectively, in both Aag-2 and C6/36 HT cells (Fig. 1c). The abundance in the expression of the two 14-3-3 isoforms was similar in the two cell lines (Fig. 1c). Lastly, intracellular distribution of 14-3-3 proteins was assessed using Immunofluorescence assay as 14-3-3 protein family interacts with other proteins and may have a distinct pattern of intracellular expression [29,30]. The 14-3-3 Table 1 Templates for DsiRNA generation designed for 14-3-3 silencing. DsiRNAs were synthesised using the custom Dicersubstrate siRNA Construction Kit (IDT, USA). RNA duplex has an asymmetric design with a single 2-base 3′-overhang on the antisense strand and is blunt on the other end; the blunt end is modified with DNA bases marked in bold  proteins, in basal conditions, were in the cytoplasm, mainly with a cortical distribution (Fig. 1d).

Effect of 14-3-3 inhibitor on C6/36 HT cells phagocytosis
To investigate the role of 14-3-3 in C6/36 HT cells functions, we assessed whether 14-3-3 inhibitor affects the cell viability, and found that viability was maintained at 90% even with highest concentration used (100 μM) and prolonged incubation time (180 min) (Fig. 2). In addition, we also evaluated whether the inhibitor induced disruption of protein-protein interaction with 14-3-3 had impact on phagocytosis. The treatment with 14-3-3 inhibitor in different concentrations ranging from 12.  (Fig. 3a). Comparable results were obtained in phagocytosis of FITC-E. coli, leading to a statistically significant decrease by up to 60% (t-test: t = 2.941, P = 0.0022), with 50 μM of the inhibitor (Fig. 3b).

14-3-3 inhibitor effect on the cytoskeleton organisation
To investigate the role of 14-3-3 proteins in actin cytoskeleton organisation, we assessed the expression pattern of actin and 14-3-3 in C6/36 HT cells in the presence or absence of 14-3-3 inhibitor. In medium and DMSO control cells, F-actin was localised mainly at the cell cortex, and 14-3-3 was present in the cytoplasm and near the cell membrane. At 120 min after treatment with 14-3-3 inhibitor, cells became elongated (prismatic), actin reorganised and co-located with 14-3-3 in filopodia-like structures (Fig. 6).
In contrast, Aag-2 cells transfected with DsiRNA14-3-3ε and DsiRNA14-3-3ζ dramatically underwent changes in their morphology from monolayer surface to a rounded cluster of cells, which has been described only in stress conditions (Fig. 8a) [32][33][34]. Moreover, the expression of both 14-3-3ε and 14-3-3ζ proteins was significantly Fig. 2 Effect of the 14-3-3 inhibitor on the viability of C6/36 HT cells. C6/36 HT cells were treated with various 14-3-3 inhibitor concentrations ranging from 12.5 μM to 100 μM for different times: at 60 min (a), 120 min (b) and 180 min (c), and cell viability was assessed. DMSO and medium alone were used as controls. Even prolonged treatment with 14-3-3 inhibitors, cell viability, was maintained at greater than 90%. Experiments were performed in triplicate; three independent experiments were conducted and represented as the mean ± standard error, SE Fig. 3 Effect of the 14-3-3 inhibitor on the phagocytosis of S. typhimurium-GFP and FITC-labeled E. coli in C6/36 HT cells. C6/36 HT cells pre-treated with 14-3-3 inhibitor at various concentrations were evaluated for phagocytosis of S. typhimurium-GFP (a) and FITC-labeled E. coli (b). Phagocytosis of both S. typhimurium and E. coli was significantly decreased by the treatment with 14-3-3 inhibitors in a dose-dependent manner. DMSO and medium alone, as well as phagocytosis performed at 4°C, were used as negative control. Experiments were performed in triplicate, and data from three experiments are presented as the mean ± standard error, SE Fig. 4 Inhibition of E. coli -containing phagosome formation by the 14-3-3 inhibitor in C6/36 HT cells. C6/36 HT cells were treated with 14-3-3 inhibitor in different concentrations and evaluated for the ability to form phagosome with pHrodo green E. coli bacteria. Phagosome formation (a, b) was significantly reduced by the treatment with inhibitors in dose-dependent manner (c) Representative histograms of phagosome formation by untreated C3/36 HT cells on ice (left, negative control, 4°C), treated with 12.5 μM of 14-3-3 inhibitor (middle), and untreated C6/36 HT cells at 34°C, in medium (right, positive control). Experiments were performed in triplicate and represented as the mean ± standard error, SE. Representative data from three independent experiments are shown reduced to 50 and 75%, respectively (t-test: t = 30.75, P = 0.0011) (Fig. 8b, c).

Discussion
The members of the 14-3-3 family are 28-30 kDa acidic proteins highly conserved in a wide range of organism and tissues [35]. The 14-3-3 proteins interact as adapters, activators and repressors with several other protein partners, regulating signalling pathways in a vast array of processes such as apoptosis, cell signalling and cytoskeletal organisation [11,36,37].
Previously, two isoforms of 14-3-3ε and 14-3-3ζ have been shown to be present in insects such as Drosophila, B. mori and Ae. aegypti [9,12,13,38]. In this work, we documented the expression and function of Aeae14-3-3ε and Aeae14-3-3ζ in C6/36 HT cells derived from Ae. albopictus and Aag-2 cells derived from Ae. aegypti as model systems from insect vectors of human diseases. We found that 14-3-3 proteins were localised in the cytoplasm in Aedes spp. cells in accordance to those observed in Ae. aegypti midguts [13]. This cellular distribution is similar to those observed in cells of several tissues of other insects, such as B. mori [38] and Pieris rapae [36]. This is important since on the isoforms that are present in different cellular locations under certain cellular conditions [29].
To investigate functions of 14-3-3 in C6/36 HT cells, we used a membrane-permeable specific inhibitor, which disrupts the formation of the functional complex through interactions with their protein partners [21,22]. 14-3-3 inhibitor had not impacted on the viability of mosquito cells in the dose range tested (90% viability after 180 min), indicating that the observed functional effect in our study was not the result of cell damage. The 14-3-3 inhibitor-induced cytoskeletal reorganisation on C6/36 HT cells which led to changes in cell shape and co-localization of 14-3-3 proteins with F-actin in filopodium-like structures (Additional file 1: Fig. S1, Additional file 2: Fig. S2). In Drosophila, 14-3-3 proteins have been suggested to interact with several regulators   Effect of the 14-3-3 inhibitor on the cytoskeleton organisation of C6/36 HT cells. The intracellular distribution of 14-3-3 proteins (green), actin (red), and nuclei (blue) on C6/36 HT cells treated with 14-3-3 inhibitor, DMSO and medium were visualised by Immunofluorescence. In medium or DMSO treated cells, 14-3-3 proteins were present mainly in the cytoplasm and did not co-localized with F-actin in the cortex. In contrast, cells treated with 14-3-3 inhibitors underwent morphologic changes, e.g. elongation through reorganisation of F-actin and colocalizing with 14-3-3 proteins. Scale-bars: 5 μm of the actin cytoskeleton and, especially, the actin depolymerising factor cofilin, which is required for the efficient phagocytosis of bacteria [3]. Besides the interaction of 14-3-3 proteins with proteins located on actin filaments, such as cofilin, interactions with other proteins has also been documented with its regulatory kinase, LIM-domain protein kinase 1 [37] and other proteins involved in attachment of the cytoskeleton to the plasma membrane and to the extracellular matrix [39,40]. In mice astrocytes, the presence of 14-3-3γ was observed to be close to actin filaments as ischemia and apoptosis induced changes in the binding of 14-3-3 to F-actin [40].
Phagocytosis is a rapid-acting immune mechanism of insect haemocytes representing a primary defence line that limits microbial infection [42][43][44]. A role of 14-3-3 proteins in phagocytosis in Drosophila was already documented [3,9,45]. C6/36 HT is a plasmatocyte-like cell line derived from Ae. albopictus that has been used as an experimental model for phagocytosis [46,47]. We  observed the reducing on S. typhimurium and E. coli uptake by C6/36 HT cells under treatment with the 14-3-3 inhibitor, indicating that 14-3-3 proteins participate in the internalisation (uptake) in the phagocytosis of Grampositive and Gram-negative bacteria. Considering that diminution of phagocytosis was indistinct for both Gram-positive and Gram-negative bacteria, the mechanism is probably independent of the receptor, and 14-3-3 proteins play a role in phagocytosis in a step after recognition of the target particle. Together, these results showed that 14-3-3 inhibitor is efficient in blocking the function of 14-3-3 proteins in mosquito C6/36 HT cells and causes the inhibition of phagocytosis of bacteria, indicating the participation of this protein in phagocytosis.
The exocyst complex is a molecular complex required for endo-and exo-cytosis, which is implicated in phagocytosis on Drosophila. The exocyst complex formation is an important step during phagosome biogenesis through the formation of the early phagosome [48]. We used pHrodo E. coli and S. aureus bacteria as useful markers of phagolysosome development because they become fluorescent when they are in the acidic environment of the phagolysosome [9,28,31,49]. In this study, C6/36 HT cells treated with the 14-3-3 inhibitor showed a significant decrease in the percentage of phagocytosis and phagolysosome formation (pHrodo bright cells), indicating the involvement of 14-3-3 proteins in phagosome maturation likely in the fusion of phagosome with lysosomes at late stage of phagocytosis [50,51], in addition to involvement of 14-3-3 proteins at early stage of phagocytosis -bacterial uptake [48,50]. These results are consistent with observations in haemocytes from Drosophila 14-3-3ɛ null mutants, where bacterial internalisation, phagosome formation and maturation were also disrupted [9]. One evidence of the 14-3-3 participation in phagocytosis was the identification of 14-3-3 protein in latex bead-containing phagosomes in J774 mouse macrophage-like cell line [51]. There is also evidence for both 14-3-3ε and 14-3-3ζ participation during phagocytosis of E. coli and S. aureus [3,9]. Here, we extended the role of 14-3-3 proteins beyond phagocytosis as our results suggest that the functional impairment of 14-3-3 proteins led to defects in phagolysosome formation.
We provided an additional line of support for the role of 14-3-3 proteins in phagocytosis through silencing expression of 14-3-3 genes at the mRNA level using DsiRNAs. RNAi mechanism depends on a molecular pathway that is triggered by long exogenous double-strand RNA (dsRNA) in the cell. Dicer-2 (Dcr2) recognises and cleaves the dsRNA into small interfering RNAs (siRNAs), which usually 21 bp initiating the RNAi pathway. The siRNA associated with Dcr2 are loaded into a multi-protein RNA-induced silencing (RISC), where it is processed (cleaved) [52][53][54]. We used a 27-mer instead of 21-mer of siRNAs and took advantage of the link between Dicer and RISC systems for silencing 14-3-3ε and 14-3-3ζ transcripts, as this approach can boost silencing efficiency by ca. tenfold compared with traditional siRNAs. We found very different outcomes in C6/36 HT and Aag-2 cells regarding morphology and protein expression after silencing 14-3-3 proteins. Aag-2 cells [55], showed a significant decrease in the 14-3-3ε and 14-3-3ζ protein expression and marked morphological changes with 14-3-3 DsiRNA silencing, in contrast to C6/36 HT cells that did not show changes in morphology nor significant decrease in 14-3-3 protein production. It has been reported that C6/36 HT cells exhibited inefficient Dcr2 cleavage of long dsRNAs or DsiRNAs while those Aag-2 cells were capable of siRNA processing. The defective expression or function of Dcr2 in C6/36 HT cells might explain the ineffective silencing likely from preventing activation of the RNAi machinery [20,56,57].
It is important to note that there are some differences between C6/36 HT cells derived from Ae. albopictus larvae whole cells [58,59] and Aag-2 derived from Ae. aegypti embryonic cells [60]. In addition, the C6/36 HT cells have been adapted to grow at 34°C (they grew at 28°C when first isolated) [61,62]. Nevertheless, filopodia-like structures have been observed in C6/36 HT cells that were infected with Japanese encephalitis virus (grew at 28°C) [63]. In addition, the Aag-2 cells grew as attached fibroblast-like cells in monolayer [32] and tended to aggregate especially when cultured for a long term [32] or infected with Rift Valley fever virus [64]. The similar aggregate was likely observed in the Aag-2 cells transfected with DsiRNA in our study.
In summary, DsiRNA treatment effectively decreased 14-3-3ε and 14-3-3ζ scaffold protein expression in Aag-2 cells, which resulted in cytoskeleton reorganisation and decreased phagocytosis and phagosome maturation. Our results demonstrated the involvement of 14-3-3 proteins in phagocytosis, a central mechanism of innate immune responses in mosquitoes. As DsiRNA mediated silencing 14-3-3ε had a bigger impact compared to 14-3-3ζ, further investigations are needed to detail differential molecular and cellular mechanisms for 14-3-3ε and 14-3-3ζ, including the identification of proteins partners connecting with the actin cytoskeleton and signalling pathways involved.